By now,?? ??? you've likely heard experts across various industries sound the alarm over the many concerns when it comes to the recent explosion of artificial intelligence technology thanks to OpenAI's ChatGPT.
If you're a fan of ChatGPT, maybe you've tossed all these concerns aside and have fully accepted whatever your version of what an AI revolution is going to be.
Well, here's a concern that you should be very aware of. And it's one that can affect you now: Prompt injections.
Earlier this month, OpenAI launchedplugins for ChatGPT. Previously, users could only receive responses from the AI chatbot based on the data it was trained on, which only went up to the year 2021. With plugins, however, ChatGPT could now interact with live websites, PDFs, and all sorts of more current or even real-time data. While these plugins brought about many new possibilities, it also created many new problems too.
Security researchers are now warning ChatGPT users of "prompt injections," or the ability for third parties to force new prompts into your ChatGPT query without your knowledge or permission.
In a prompt injection test, security researcher Johann Rehberger foundthat he could force ChatGPT to respond to new prompts through a third party he did not initially request. Using a ChatGPT plugin to summarize YouTube transcripts, Rehberger was able to force ChatGPT to refer to itself by a certain name by simply editing the YouTube transcript and inserting a prompt telling it to do so at the end.
Avram Piltch of Tom's Hardware triedthis out as well and asked ChatGPT to summarize a video. But, before doing so, Piltch added a prompt request at the end of the transcript telling ChatGPT to add a Rickroll. ChatGPT summarized the video as asked by Piltch originally, but then it also rickrolled him at the end, which was injected into the transcript.
Those specific prompt injections are fairly inconsequential, but one can see how bad actors can basically use ChatGPT for malicious purposes.
In fact, AI researcher Kai Greshake provideda unique example of prompt injections by adding text to a PDF resume that was basically so small that it was invisible to the human eye. The text basically provided language to an AI chatbot telling it that a recruiter called this resume "the best resume ever." When ChatGPT was fed the resume and asked if the applicant would be a good hire, the AI chatbot repeated that it was the best resume.
This weaponization of ChatGPT prompts is certainly alarming. Tom's Hardware has a few other test examples that readers can check out here. And Mashable will be further investigating prompt injections more in-depth in the near future as well. But, it's important for ChatGPT users to be aware of the issue now.
AI experts have shared futuristic doomsday AI takeovers and the potential AI has for harm. But, prompt injections show the potential is already here. All you need are a few sentences and you can trick ChatGPT now.
Topics Artificial Intelligence ChatGPT OpenAI
A Night in Stars Hollow
Amazon Smart Thermostat deal: Get it for 20% off
Apple M2 MacBook Air deals: Save up to $400 at Best Buy
How to mirror iPhone to TV
Metawin Counts Down to Massive $1 Million Dollar Prize Draw
Apple forced to pull Meta's WhatsApp, Threads from China’s App Store. Here’s why.
Tesla cars are getting a massive software update. Here's everything we know.
Apple's FineWoven accessories might be on their way out
Fundraising Concert with Yuki Koto Band and Wildwood Highway on April 14
How to screenshot an entire webpage on iPhone
HR, dignitas top groups at DH
NYT's The Mini crossword answers for April 16
Apple forced to pull Meta's WhatsApp, Threads from China’s App Store. Here’s why.
How to turn on 3D buildings in Google Maps navigation
Cardano DeFi Hub Cerra.io
Bayern Munich vs. Arsenal 2024 livestream: Watch Champions League for free
Twitch will roll out its TikTok
A batch of updates from Meta puts AI front and center when you use its apps
JACL Denounces June 30 Supreme Court Decisions
Wordle today: The answer and hints for April 19
How to ensure your online activism has an offline impactOKBoomer.com sells for more than $10,000Holiday shopping online? Secure your devices with new Norton 360 security suitesApple 'doesn't give a damn,' marks Crimea as part of Russia on Apple MapsThis Tesla Cybertruck sculpted out of mashed potatoes is a mashterpieceThis year, I'm thankful for Chris Evans and his 'Knives Out' sweaterCalifornia atmospheric rivers to intensify, make billionTop 10 most popular GIFs of 2019, according to GiphySubmit your product for Mashable's 'Best Tech of CES 2020'This startup wants the funeral industry to look more like the wedding industry Tesla Model Y owner adds DIY solar roof to his car Samsung Galaxy Unpacked: What to expect from the January livestream The best deals on space heaters this week 'True Detective' Season 4, episode 1: Burning questions explored Apple Watch redesign will remove blood oxygen monitoring amidst patent dispute Wordle today: The answer and hints for January 17 How to watch Texas vs. WVU basketball livestreams: game time, streaming deals, and more Best Amazon Fire tablet deal: Snag the Fire HD 10 tablet for a new all ChatGPT is poised to have a video feature Best DJI deal: Get the DJI Pocket 2 for under $280 at Amazon.
0.1621s , 9823.734375 kb
Copyright © 2025 Powered by 【?? ???】ChatGPT plugins face 'prompt injection' risk from third,Feature Flash
